Critical role of particle/polymer interface in photostability of nano-filled polymeric coating.
ZnO nanoparticles can function as a photocatalyst or a photostabilizcr, depending on the UV exposure conditions. A hypothesis is proposed that the polymers in the vicinity of the ZnO/PU interface are preferentially degraded or protected, depending on whether ZnO nanoparticles act as a photocatalyst or a photostabilizer in the polymers. This study clearly demonstrates that the particle/polymer interface plays a critical role in the photostability of nano-filled polymeric coatings.
Keywords ZnO nanoparticles, Interface, Photodegradation, Accelerated test, Polyurethane, Waterborne, Polymeric coatings, UV irradiation, NIST SPHERE
Nanoparticle-filled polymeric coatings have attracted great interest in recent years because incorporation of nanoparticles in the form of spheres, tubes, rods, and other shapes can significantly enhance mechanical, electrical, optical, thermal, and antimicrobial properties of the coatings. This enhancement can be attributed to (1) unique nanoparticle properties (e.g., mechanical and electrical), (2) small size, and (3) dramatically increased volume fraction of interfacial area. A conservative estimate of the interfacial volume fraction for 10 nm radius spherical fillers at a loading of 1% by volume fraction is about 5% of the polymer composite, in contrast to microscale fillers for which the interfacial volume at the same loading is negligible.1 Because of the large volume fraction of the interfacial area, the properties of the interfacial region play a critical role in dispersion quality, stress transfer,2 charge transfer, and long-term durability of a nano-filled polymeric coating system. Therefore, optimization of the nanoparticle/polymer interfacial properties is a subject of extensive research in polymer nanocompositcs. One simple way to accomplish this is by modifying the nanoparticle surfaces via physical adsorption with surfactants or polymers to render them more compatible with the polymer matrix, (3) or by chemically functionalizing the nanoparticle surfaces so that covalent bonds can be formed between the nanoparticles and polymer matrix. (1), (2), (4-7) Compared to physical adsorption, the formation of covalent bonds is generally more effective for stronger interfaces and stress transfer. However, structural changes during chemical functionalization can cause unwanted defects in the nanolillers. (8)
In addition to optimizing the performance of a nano-filled polymer, it is imperative to characterize and understand how interfacial structures/properties affect long-term performance of these nano-filled polymeric materials. It is well-established that the particle/polymer interface (including interphase) is the region in the vicinity of the particle surface, where polymer properties are altered as compared to the bulk. (9-11) The size scale of the interface can be as small as one molecule when the polymer chain is perturbed in the immediate vicinity of the particle surface ([approximately equal to] 10 nm). (12), (13) However, it can also extend to hundreds of nanometers in some systems, depending on the surface chemistry of particles, the dynamics of matrix polymer chains, and the surface structure of the particles. (14) For example, a previous study reported that (15) multiple polymer layers were observed to sheath the multiwalled carbon nanolubes (MWCNT) that were pulled out from the polycarbonate matrix. After the outer layer of the polymer ([approximately equal to] 160 nm thick) was removed with an AFM tip, a thin layer ([approximately equal to] 40 nm thick) of polymer was still found adhered to the nanotube. This study suggested that two distinct interfaces--a MWCNT-inner polymer layer interface and an inner polymer layer-outer polymer layer interface--existed in this MWCNT-polymer system. This result demonstrated that, depending on materials properties and processing conditions, the nanoparticle/polymer interface can become very complex. Chemical functionalizalion and presence of dispcrsants/surfactants can substantially alter the original nanoparticle surface. Likewise, the structure, the thermodynamics, and the chemical reactivity of the matrix near the interface also are altered because polymers are confined within geometries whose characteristic dimensions approach molecular dimensions. (12) The polymer/nanoparticle interface can contain surface contaminants, unreacted or partially reacted low-molecular weight molecules, and processing additives. The interface is further affected by processing conditions, which may cause chemical reactions, species diffusion, volumetric changes, and residual stresses. (16-18) Each of these phenomena can vary in scale and occur concurrently at and near the interface. Therefore, the structure and volume fraction of the interface have shown a profound effect on mechanical properties and thermal stability of polymer nanocomposites and polymeric coatings. (1), (2), (6)
Due to the small size of nanoparticles, characterizing and mapping the nanoparticlc/polymer interface is extremely challenging. Atomic force microscopy (AFM) is well-established as a powerful, nondestructive technique that can resolve nanoscale features on the surface of a material. However, for a typical filled polymeric coating system, a thin "clear coating layer" where the fillers are absent is often observed on the outer coating surface. (19), (20) The thickness of this clear coating layer can be on the order of a few microns and depends on the total film thickness, the pigment particle size, and the wettability and dispersion of particles in the coatings. (20) This clear layer makes it difficult to directly observe the nanoparticle/polymer interface from the coating surface using conventional AFM. An alternate way to observe the nanoparticle/polymer interface is to image the fractured surface of the nano-filled polymer films. However, the nanoparticlc/polymer interface could be covered by polymer on the fractured surface, particularly for systems having strong intcrfacial bonding.
In this study, we investigated the role of nanoparticle/polymer interface in the photostability of polymeric coatings. A nano-ZnO-filled waterborne polyurethane (PU) coating was selected as a model system. The nanoparticle/polymer interfacial properties were varied by changing surface treatment, size, and loading of the ZnO nanoparticles. The influence of these variations on chemical, thermo-mcchanical, and morphological properties of the PU coatings during ultraviolet (UV) irradiation was evaluated. The interfacial regions of the PU/ZnO films before and after UV exposure were characterized by AFM-based techniques. Results showed that the interfacial regions in the ZnO/PU films were successfully detected by using a combination of tapping mode AFM and a novel electric force microscopy (EFM) technique. The nature of the particle/polymer interfacial regions played a critical role in photostability of the UV-exposed ZnO/PU films. In addition, the results have shown that the ZnO nanoparticles can act as a photocatalyst or a photostabilizer, depending on the UV exposure conditions, and the degradation behavior near/at the interfaces of the specimens exposed to these different conditions is discussed.
Materials and specimen preparation
A commercial waterborne dispersion of PU (Bayhydrol 110, Bayer Material Science) and a series of waterborne dispersion of ZnO nanoparticles (BYK) Additives & Instruments) were selected for the preparation of nano-ZnO-filled PU films. The PU coating was a one component, anionic dispersion of an aliphatic polyester urethane resin in water/n-methyl-2-pyrrolidone solvent. The average diameters of the ZnO nanoparticles were 20, 40, and 60 nm, and their specific surface areas were 54, 33, and 18 [m.sup.2]/g, respectively (as provided by the manufacturer). Two types of surface treatment, designated as BYK-treated and LP-treated, were used for 40-nm diameter ZnO nanoparticles. The 20 and 60-nm diameter nanoparticles were all BYK-treated. The surface treatments of these nanoparticles were carried out during manufacturing to provide good dispersion and compatibility with the polymer matrices. However, specific information on the surface modification process was not disclosed by the manufacturer. It is noted that untreated ZnO nanoparticles are highly aggregated and difficult to fully disperse in water, and modifying the surface of these nanoparticles with different functional groups in the laboratory to achieve good dispersion is very challenging.
ZnO nanoparticle/PU composite films (hereafter referred to as ZnO/PU films) were prepared by mixing the PU dispersion with different loadings of ZnO nanoparticles using a mechanical stirrer (Dispermat, VMA) at 315 rad/s (3000 rpm) for 20 rain. Three ZnO nanoparticle loadings, 1, 2, and 5% (based on mass of the solid PU), were used. After degassing for 1 h in vacuum, the mixture was then applied to the substrates. Thin films with thickness of approximately 3-5 [micro]m were prepared by spin coating onto calcium fluoride (Ca[F.sub.2]) substrates. Thick films having a nominal thickness of approximately 100 [micro]m were prepared by drawdown technique on a glass substrate that was pretreated with a release agent. All films were dried overnight under ambient conditions, followed by an oven post-curing at 150[degrees]C for 10 min. Free-standing ZnO/PU films were obtained by removing the thick films from glass substrates. In addition, PU films without ZnO nanoparticles also were prepared for comparison. The low absorption in the visible spectrum of all ZnO/PU films as measured by UV-visible spectroscopy indicates that the ZnO nanoparticles are well-dispersed in these systems. (21)
High UV radiant laboratory exposure
The UV exposure of ZnO/PU films was mainly conducted on the NIST simulated pholodegradation via high energy radiant exposure (SPHERE), a 2-m diameter integrating sphere-based weathering chamber. (22), (23) This weathering device utilizes a microwave-powered, mercury are lamp system that produces a collimated and highly uniform UV flux of approximately 480 W/[m.sup.2] at 100% power in the spectral range between 290 and 400 nm. Within the sample chambers, the relative humidity (RH) and temperature can be controlled precisely and individually. The emission spectra of the light source of the SPHERE and other details about the SPHERE technology can be found in reference (22). In this study, ZnO/PU films coated on Ca[F.sub.2] were placed in two 17-window sample holders, and exposed to UV radiation at 45 [+ or -] 0.5[degrees]C, 0% RH [+ or -] 5% RH and 45 [+ or -] 0.5[degrees]C, 75% RH [+ or -] 5% RH (hereafter referred to as 45[degrees]C, 0% RH and 45[degrees]C, 75% RH, respectively). The exposed specimens were removed at specific intervals for characterization. To study the UV exposure conditions on the photocatalytic effect of nanoparticles, ZnO/PU films were also exposed to a UV/Ozone (UVO) cleaner (Jetlight Company, Inc.). The main wavelengths of the light emission of this device are 254 and 185 nm.
Fourier transform infrared spectroscopy
Chemical degradation of the ZnO/PU films coated on Ca[F.sub.2] was measured by transmission Fourier transform infrared spectroscopy (FTIR) using a PIKE auto-sampling accessory (PIKE Technologies). This automated sampling device allows efficient and rapid recording of the transmission FTIR spectra. As the exposure cell was mounted precisely on the auto-sampler, errors due to sampling variations at different exposure times are minimized. The auto-sampler accessory was placed in an FTIR spectrometer compartment (Thermo-Nicolet Nexus 670x1) equipped with a liquid nitrogen-cooled mercury cadmium telluride (MCT) detector. Spectra were recorded at a resolution of 4 [cm.sup.-1] and were averaged over 128 scans. The peak height was used to represent IR intensity, which was expressed in absorbance units.
Atomic force microscopy and electric force microscopy
A Dimension 3100 AFM (Veeco Metrology) was used to image the morphology and the microstructure of the ZnO/PU films before and after UV exposure. The AFM was operated in the tapping mode using commercial silicon probes (Olympus AC-160TS) with a resonance frequency of approximately 300 kHz. The set-point ratio (the ratio of set-point amplitude to the free amplitude) ranged from 0.60 to 0.80. Successive AFM measurements were performed at essentially the same locations on a specimen to facilitate the study of morphological changes occurring in the films as a function of UV exposure.
Two specific EFM techniques applied here were scanning surface potential microscopy (SSPM or Kelvin probe force microscopy as it is often referred to in the literature) and scanning conductance microscopy (SCM). Both techniques were conducted on the same AFM using a conductive Pt-coated EFM probe (Olympus AC240-TM) and a two pass scanning technique. During the first pass, the topography and main-phase images were acquired in normal tapping mode. During the second pass, the conductive AFM probe was raised above the sample by a fixed distance (10 nm here) and the surface was rescanned. Both AC and DC voltages were simultaneously applied to the probe for SSPM, while only a DC voltage was applied for SCM. Simultaneously, the surface potential image and the EFM lifted-phase shift of the vibrating AFM probe (referred as EFM phase) were recorded in these two techniques, (24)
Dynamic mechanical thermal analysis
Storage modulus (E') and glass transition temperature ([T.sub.g]) of the ZnO/PU films were measured on an RSA III (TA Instruments) dynamic mechanical thermal analyzer (DMTA). Measurements were performed from -100 to 150[degrees]C at a temperature ramp of 3[degrees]C/min, with a frequency of 1.0 Hz and a strain of 0.5%. [T.sub.g] was determined from the maximum of the tan-[delta] peak, and was the average of three measurements.
Electron paramagnetic resonance spectroscopy
Electron paramagnetic resonance (EPR) analysis was conducted to quantify the photoreactivity of ZnO nanoparticles using radical spin traps containing 3-aminoproxyl (AP) (Aldrich) at 500 [micro]mol/L in deionized water and stock preparations (0.2 g/L) of metal oxide suspensions in deionized water. For each spin trap experiment, a new mixture of spin trap stock solution and ZnO suspension was prepared to monitor the photoinitiated radicals under UV irradiation at ambient temperature. Aliquots (50 [micro]L) of the spin trap/ZnO mixture were placed into EPR capillary tubes in the EPR cavity and irradiated in situ with a Xe arc lamp at 500 W. All EPR spectra were recorded at ambient conditions with a Bruker Elexsys E500 EPR spectrometer. Measurements were carried out at a resonant frequency of 9.38 GHz and a microwave power of 0.6-10 mW. The time constant, conversion time, sweep time, and signal receiver gain were adjusted to obtain optimum signal resolution. EPR spectra were recorded at regular time intervals during the UV irradiation period. Control experiments were carried out to ensure that observed EPR signals did not arise from photolysis or oxidation products of the spin traps themselves as well as to monitor the EPR signal stability in the dark and under illumination.
[FIGURE 1 OMITTED]
Results and discussion
Thermo-mechanical properties of nano-filled coatings and their relationship with interfacial interactions
One of the thermo-mechanical properties of a polymer that can be greatly affected by nanofillers is the glass transition temperature ([T.sub.g]). (1), (25-27) It has been reported that the [T.sub.g] of a polymer nanocomposite can increase or decrease, depending on the attractive or repulsive interactions between the nanoparticle surface and the polymer matrix. (25-27) If the particle surface is attractive or compatible with the polymer, [T.sub.g] will increase; if it is repulsive or dewetting, [T.sub.g] will decrease. The mechanism for these phenomena is still not clear.
Figure la displays DMTA loss tangent (tan-[delta]) as a function of temperature for ZnO/PU films with various loadings of 20 nm ZnO nanoparticles. Multiple peaks are observed in the plots for both pure PU and ZnO/PU films, corresponding to the multiple phases in PU. The most evident peak near 100[degrees]C is attributed to the [T.sub.g] of the polymer hard segments in this one-component PU system, and will be used for comparing [T.sub.g] values of PU films prepared with different conditions of ZnO nanoparticles. A substantial and progressive increase in [T.sub.g] with increasing ZnO nanoparticle loading is observed (94,114,123,127[degrees]C for 0,1, 2, and 5% ZnO/PU films, respectively), along with a sequential drop in the intensity of the maximum tan-[delta] peak. This observation suggests that attractive forces exist at the nano-ZnO/PU interface, leading to a reduction in the mobility of the PU polymer chains in the hard segments. Recent molecular dynamics simulations showed that the relaxations of chain segments in the immediate vicinity of the nanoparticles are slow compared to those away from the particles. (28), (29) Note that the ZnO nanoparticles used in this study were modified with wetting agents/surfactants by the manufacturer for the dispersion and designed to be compatible with the waterborne polymer matrix. Without more detailed knowledge of the surface chemistry of the modified ZnO nanoparticles, it is difficult to develop an accurate mechanism on how the PU molecules interacted with the nanoparticle surfaces. Secondary forces such as hydrogen bonding, dispersion, or electrostatic forces might be the main reasons for these interfacial interactions. The higher [T.sub.g] with a higher ZnO concentration can also be attributed to the increased volume fraction of interfacial area, and hence the overall increased interfacial interactions between nanoparticles and polymers, because no substantial difference in the dispersion of the nanoparticles was observed for ZnO/PU films at different loadings (data not shown). The increased interfacial volume may also account for the enhanced storage modulus (E') at temperature above 50[degrees]C for the systems with higher ZnO nanoparticle loadings (Fig. lb).
Another method of altering particle/polymer interfacial interactions in nano-filled polymeric coatings is to utilize nanoparticles having different sizes while maintaining the same loading. Figure 2a displays tan-[delta] as a function of temperature for ZnO/PU films with ZnO nanoparticles having diameters of 20, 40, and 60 nm at a 5% loading. As a comparison, the tan-[delta] data for unfilled PU film is also presented. A substantial increase in [T.sub.g] with the addition of ZnO nanoparticles is observed for all three particle sizes, and the increase in [T.sub.g] is greatest for the 20 nm particles (+33[degrees]C) and smallest for the 60 nm particles (+17[degrees]C). Note that surface modifications for the different sizes of nanoparticles are the same. Assuming that the nanoparticle dispersion was similar for all these nanoparticles, the interfacial interactions between the nanoparticles and the polymers are mainly determined by the overall interfacial volume fractions of the nanocomposite, which are proportional to the specific surface area of the nanoparticles. Based on the observed relationship between the [T.sub.g] shift and the particle size, we suggest that the interfacial interactions for the 20-nm diameter-filled ZnO/PU system are greater than those for the 60 nm filled system. Changes in the moduli of ZnO/PU films containing different sizes of particles also follow the same trend as changes in the [T.sub.g] (Fig. 2b).
[FIGURE 2 OMITTED]
Surface modification of the nanoparticles can also alter interfacial properties resulting in a shift in [T.sub.g]. Figure 3a shows the tan-[delta] as a function of temperature for ZnO/PU films containing two types of surface-modified 40 nm-sized ZnO nanoparticles at 5% loading. As can be seen, the [T.sub.g] values corresponding to the hard segments of polymers are similar ([approximately equal to]118[degrees]C) for both systems; however, the [T.sub.g] corresponding to the soft segments of polymers is much higher for LP-treated system (-35[degrees]C) than that of the BYK-treated system (-62[degrees]C). The peak also appears more pronounced for the former. Different surface modifications of ZnO nanoparticles also lead to changes in storage moduli, especially at low temperatures (<-0[degrees]C) for the two filled systems (Fig. 3b). The above results clearly indicate that the macroscopic thermomechanical properties, such as [T.sub.g], can be used as an indication of the particle/polymer interactions in a nano-filled polymeric coating.
Detecting embedded nanoparticle/polymer interface using A FM-based techniques
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
The effects of ZnO nanoparticles on the surface morphology and niicrostructurc of the ZnO/PU films ([approximately equal to]5 [micro]m thick) spin-coaled on Ca[F.sub.2] are shown in Figs. 4 and 5. As seen in Fig. 4a, the surface of the pure PU film is essentially smooth, showing few topographic features except for some defects due to sample preparation. After incorporation of ZnO nanoparticles, the surface roughness of the films increases with increasing ZnO content (Fig. 4b) and some protruding regions and small holes are also observed. These protruding regions initially were suspected to be aggregates or agglomerates of nanoparticles; however, high magnification AFM images (Fig. 5) and in particular the AFM phase images show that only a few nanoparticles--not large aggregates of nanoparticles--are observed on the surface of these protruding regions. The near absence of nanoparticles on the surface suggests the existence of a "clear coating layer" on the outer surface of the ZnO/PU films. (19), (20) As mentioned previously, the clear layer is common for pigmented polymeric systems. The formation of this layer is probably driven by the low-surface energy of polymers and the gravity of the pigments. (19), (20), (30) The thickness of the clear layer depends on the total film thickness, the pigment particle size, and the wettability and dispersion of particles in the polymer matrix. (19), (20) The presence of a clear layer in the ZnO/PU composite films has strong implications on the photodegradation of this nano-filled material.
[FIGURE 5 OMITTED]
Meanwhile, more holes are observed on the surface of a ZnO/PU film with a higher ZnO loading (Fig. 5). The depths of these holes range from a few nanometers to tens of nanometers. These holes indicate that the coalescence of the film was interrupted by the presence of ZnO nanoparticles, leading to a nonpcrfect film formation for the studied one-component waterborne PU system. (31-33) The phase contrast image shown in Fig. 5 reveals that the PU films have a two-phase heterogeneous nanostructure with bright domains about 10-20 nm in diameter. This nanostructure is similar to that observed for two-component solvent-borne PU systems, (34), (35) and can be attributed to the nanophase separation of the hard and soft phases in the PU. Note that these nanostructural domains are independent of their surface topography. No distinct differences in AFM phase contrast are observed for the films with different ZnO loadings. Regardless of whether the regions are inside the holes or on the protrusions, the phase contrast and the size of the bright domains are similar. A number of particles that show relatively brighter contrast in the phase images are ZnO nanoparticles; the apparent diameter of these particles are around 35-45 nm. Considering the AFM tip effect, it is reasonable to suggest that these particles are monodispersed ZnO nanoparticles. Interestingly, some nanoparticles are observed to be surrounded by an interfacial layer in darker phase (see inset in Fig. 5). The thickness of this interfacial layer is approximately 15 nm. Since the phase contrast obtained by AFM lapping mode is derived from property differences (such as mechanical or/and chemical property disparities) between different domains, (36-39) we postulate that this layer has different properties from those of the polymer matrix, which likely results from the interfacial interactions between the ZnO nanoparticle surfaces and the PU molecules. This observation is consistent with the increase in [T.sub.g] after incorporation of ZnO nanoparticles into the PU matrix.
Note that, due to the presence of clear coating layer, only a very few nanoparticlcs close to the top surface of the 5 [micro]m-thick films could be detected with tapping mode AFM and show the discernible interfacial regions. In order to obtain more information on the nanoparticle/PU interface, we prepared thinner PU/ ZnO films ([approximately equal to] 3 [micro]m thick) and employed SSPM and SCM, techniques based on long-range electrostatic forces between the probe and the sample, to map the surface and subsurface nanostructures of these films. Figures 6a-6d represent height image, surface potential image, EFM lifted-phase image, and EFM main-phase image, respectively, of the ZnO/PU films containing 5%, BYK-treated 20 nm-sized ZnO particles. All images were obtained at the same location on the sample. Numerous particles are shown in the height image of the film, but essentially no features are observed from the surface potential image. Meanwhile, from both EFM lifted-phase and EFM main-phase images, we can clearly see particles distributed randomly in the polymer matrix, with slightly different contrast in the interfacial areas surrounding the particles. Compared to the height image, these phase images resolve particles and interfaces more clearly. The apparent diameters of these particles range from 40 to 60 nm. Considering the AFM tip effect and the EFM broadening effect from subsurface imaging, it is reasonable to suggest that the majority of these particles are monodispersed ZnO nanoparticles.
[FIGURE 6 OMITTED]
The different features observed in various EFM images are due to their different contrast mechanisms. As mentioned above, all these images were obtained using two-pass scanning techniques. The AFM height and EFM main-phase images were recorded during the first pass when the AFM probe was intermittingly contacted with the sample surface, while surface potential and EFM phase images were recorded during the second pass when the conductive probe was raised above the sample at a fixed distance (e.g., 10 nm) with a bias. Comparing the surface potential image with EFM lifted-phase image, the former is based on electrostatic force which is more dominated by the surface chemistry or polarity of the polymeric materials, while the latter is based on electrostatic force gradient, which is sensitive to subsurface features. The lack of feature in the surface potential image in Fig. 6b indicates that the outermost surface layer of these ZnO/PU films is dominated by a clear coating layer, composed mainly of PU molecules. The observed EFM lifted-phase contrast mainly reflects the subsurface structure,  which is attributed to the ZnO particles located in the subsurface regions. The darker phase contrast in the regions surrounding the particles indicates that interfacial regions have slightly different dielectric properties from the polymer matrix. In addition, the EFM main-phase image shown in Fig. 6d also reveals the interfacial regions around the particles. The size of this interfacial layer detected by EFM ranges from 20 to 40 nm, which is consistent with the previous reports. (12,) (15) These preliminary results suggest that, by combining different AFM-based techniques, it is possible to detect the nanoparticle/polymer interfacial regions directly from the top of the coating surface.
Role of nanoparticle/polymer interface in photostabitity of nano-filled polymeric coatings
[FIGURE 7 OMITTED]
To understand the influence of the nanoparticle/polymer interface on the long-term performance of nano-filled polymeric coatings, we varied the size, the concentration, and the surface treatment of the ZnO nanoparticles, and studied the effect of these parameters on the photodegradation rate of the ZnO/PU films. The effect of ZnO nanoparticles on morphological changes of PU films can be seen in Figs. 7 and 8. Films of both pure PU (Fig. 7) and 2% ZnO/PU (Fig. 8) were exposed on the SPHERE at 45[degrees]C, 0% RH for 2 months. Multiple scan sizes of 50 x 50 [micro]m, 5x5 [micro]m, and 1 x 1 [micro]m are shown. For the pure PU after exposure, a number of randomly distributed pits are observed, ranging from a few nanometers to 150 nm in depth, and tens of nanometers to over 10 [micro]m in diameter. These isolated pits are similar to those observed for the solventborne PU (40) and acrylic melaminc films (41) following UV exposure. Some small holes are observed on the surface of the 2% PU/ZnO film before exposure, as shown in the 5x5 [micro]m scan size (Fig. 8a). As discussed earlier, these small holes are likely due to the incomplete film coalescence in the presence of ZnO nanoparticles. After UV exposure, numerous deeper holes are observed, ranging from 20 to 40 nm in depth and 150 to 250 nm in diameter. Compared to those observed in the pure PU, these pits are generally smaller, but deeper and greater in number, spreading over the entire surface. In some locations, much deeper and larger holes are observed. The AFM height and phase image of one such hole is shown in Fig. 8b. A number of nanoparticles are clearly seen inside the hole, suggesting that the polymers next to the nanoparticles have preferentially degraded as compared to the bulk polymer matrix. In other words, local accelerated degradation of the polymer has taken place in the interfacial region near the nanoparticle surface. The numerous small holes most likely result from local acceleration of the polymers near the well-dispersed nanoparticles. In regions having abundant nanoparticles, and if these particles catalyze the photoreactions, the degradation rate would be higher, resulting in deeper and larger holes.
[FIGURE 8 OMITTED]
[FIGURE 9 OMITTED]
[FIGURE 10 OMITTED]
The photoreactivity of ZnO nanoparticles is confirmed by EPR spin trap measurements as shown in Fig. 9. The results clearly indicate that the photoreactivity of the ZnO nanoparticles has a medium level of photoreactiviiy, falling between the less reactive [Al.sub.2][O.sub.3]-coated pigmentary Ti[O.sub.2] and more reactive nano-sized Ti[O.sub.2]. Further, the photocatalytic effect of ZnO nanoparticles on polymer degradation is also reflected by the chemical and mechanical degradation of the ZnO/PU films during SPHERE exposure. Figures 10 and 11 show the effect of the ZnO nano-particles on chemical degradation of PU films exposed on SPHERE at 45[degrees]C, 0% RH. The FTIR difference spectra of films of pure PU and 5% BYK-treated 20 nm-sized ZnO/PU for different UV exposure times are presented in Fig. 10. The spectra for eight exposure times--4, 7, 14, 18, 21, 25, 32, and 39 days--are displayed sequentially for each specimen (Fig. 10). Little change is observed for pure PU films. However, substantial spectral changes are seen for the ZnO/PU films and the magnitude of these changes increase as a function of exposure time (Figs. 10, 11). Note that the specimens exposed to the same temperature/RH condition without UV light show little degradation. The degradation and chain scission reactions are represented by the decrease in the intensity of C[H.sub.2] asymmetric stretching at 2928 [cm.sup.-1], ester C=O stretching band at 1732 [cm.sup.-1], amide II band (mostly due to NH banding) at 1528 [cm.sup.-1], C[H.sub.2] wag and ester C-O-C stretching band at 1250 [cm.sup.-1], and ester C-O-C asymmetric stretching band at 1041 [cm.sup.-1]. (40), (42) As shown in Fig. 11, rates of chain scission are accelerated with an increase in the ZnO nanoparticle loadings. This trend is in agreement with the observed changes in thermo-mechanica] properties shown in Fig. 12, which clearly displays that the loss in [T.sub.g] and storage modulus of the ZnO/PU films increase with increasing ZnO nanoparticle loading.
The above morphological, chemical, mechanical, and photoreactivity results are self-consistent, and clearly demonstrate that the ZnO nanoparticles used in this study had a photocatalytic effect on the degradation of PU during exposure to the SPHERE. We also observed that this photocatalytic effect is dependent on the size of nanoparticles, with smaller particles having a more profound acceleration effect (Fig. 13). Because cither an increase in the loading or a decrease in the size of the nanoparticles leads to an increase in the interfacial volume fraction, it appears that this photocatalytic effect is an interface-related phenomenon, dependent on interfacial chemistry, total interfacial area, and the structure of the interfacial region. This hypothesis is further confirmed by the evidence that different surface treatments led to different degradation rates for 5% ZnO/PU films containing 40 nm ZnO nanoparticles (Fig. 14).
[FIGURE 11 OMITTED]
[FIGURE 12 OMITTED]
[FIGURE 13 OMITTED]
However, when we changed the UV exposure device from SPHERE to a UVO cleaner, which contained unnatural radiation in short wavelengths (254 and 185 nm) along with ozone, we found that the degradation rate of pure PU became the highest among the films. As shown in Fig. 15, when the particle size was fixed at 20 nm, films with 5% loading still degraded at a faster rate than the 1% filled system, but slower than the pure PU films. Further, all three sized nanoparticles (20, 40, and 60 nm) decelerated the degradation of PU films exposed to this exposure condition; that is, they act as a UV stabilizer. Such a different phenomenon can be attributed to the different degradation mechanisms for the specimens exposed to the different UV devices. Compared with the SPHERE, a system that produces a collimated and highly uniform UV flux of approximately 480 W/[m.sup.2] in the range from 290 to 400 nm, the UVO cleaner uses a low-pressure mercury lamp, which is a shorter wavelength UV light source consisting mainly of lines at 254 and 185 nm. Atomic oxygen is continuously generated, and ozone is continually formed and destroyed by these two UV wavelengths. The PU molecules have strong absorption at wavelengths around 200 nm, but not at those longer than 290 nm. (21) Although detailed mechanism for chemical degradation of PU in the UVO cleaner needs further study, it is reasonable to conclude that in the UVO-cleaner condition, ZnO nanoparticles reduce the degradation of polymers (act as a pholostabilizer), while in the SPHERE exposure condition, they accelerate the photodegradation of polymers (act as a photocatalyst).
[FIGURE 14 OMITTED]
[FIGURE 15 OMITTED]
[FIGURE 16 OMITTED]
In fact, the UV-absorption properly is an intrinsic attribute of semiconducting oxides such as ZnO. Due to its broadband UV absorbance from 240 to 380 nm, a layer of ZnO coating can effectively prevent the underneath polymer from photodcgradation. (43) How-ever, both photocatalytic and UV protection effects have been reported for ZnO nanoparticles filled in the polymer matrices. (21), (44-50) This can be explained by the mechanism of photocalalysis in metal oxide semicon-ductors. (50) When the semiconductors such as ZnO arc illuminated by light with energy greater than the band gap (3.4 eV), electrons will be excited from the valence band to the conduction band, leaving behind positively charged species, called holes. These excited electrons and holes then follow one of two competing pathways: (1) combining with other holes or electrons or (2) being captured by the absorbents surrounding the semiconductors such as water molecules, hydroxyl groups at the metal oxide surfaces, forming hydroxyl free radicals, and other radical species. When the former pathway is dominant, the semiconductors will show a photostabilizing effect, acting as a UV protector; otherwise, they will behave as a photocatalyst to accelerate the degradation of polymers.
Based on data shown in Figs. 11-15, we suggest that the photocatalytic effect of ZnO nanoparticles can be affected by the size, the loading, and the surface chemistry of the particles. The exposure conditions such as the short wavelength of the UV light source with the presence of the ozone could affect the photocatalytic reactions of nanoparticles with polymers as well. Our previous studies also show the RH during UV exposure is another factor for the photocatalytic effect.2 Furthermore, other researchers have found that the properties of ZnO nanoparticles in polymeric coatings are also dependent on the chemistry of the coating system.44 Therefore, it is complicated to predict and evaluate the photocatalytic or photostabilizing ability of the ZnO nanoparticles when they are added to a particular polymeric coating. All of these parameters eventually lead to the different intcrfacial interactions between ZnO nanoparticles and polymer matrix during UV irradiation. As a result, the degradation of the polymers in the vicinity of the nanoparticles will be different, depending on whether the ZnO nanoparticles behave as a photostabilizer or a photocatalyst.
Figure 16 shows an example for the case when ZnO nanoparticles act as a photostabilizer (exposure in UVO cleaner). The regions next to the particles are observed to change more slowly than the bulk polymer matrix, forming elevated domains, not holes (like features shown in Fig. 8), on the surface of the degraded film. Comparing the AFM images acquired from SPHERE exposure (Fig. 8) with those from UVO-cleaner exposure (Fig. 16), we believe that the function of nanoparticles is mainly a local effect on the polymer degradation. The polymer matrix in the vicinity of the nanoparticles is preferentially degraded when nanoparticles behave as a photocatalyst, or protected when nanoparticles behave as a photostabilizer. Such differences in the local interfacial regions ultimately will result in differences in overall performance for polymeric coatings during UV irradiation. The results of this study suggest that structure and properties of the nanoparticle/polymer interface play a critical role in photostability of nano-filled polymeric coatings. If the interface can be designed to enhance the stabilizing effect or substantially reduce the catalytic effect (e.g., free radical scavenger is preferentially adsorbed on the surface of nanoparticles or the particles are encapsulated by a layer of Si02 or SiCV [A1.sub.2.sup.3] (51)), the weathering service life of nano-filled polymeric coating systems would be optimized.
In this study, we have investigated the role of nano-particle/polymer interface in photostability of poly-meric coatings using a waterborne PU coating filled with ZnO nanoparticles during UV irradiation. The effects of the parameters that vary the particle/polymer interfacial properties, such as size, loading, surface modification of the nanoparticles, on photodegradation of ZnO/PU films were evaluated. Both tapping mode AFM and novel EFM techniques were used to char-acterize the interfacial regions of ZnO/PU films. Based on our results, the following conclusions are made:
(1) The macroscopic thermo-mechanical properties, such as [T.sub.g], can be used as an indicator of the particle/polymer interactions in nano-filled poly-meric coatings.
(2) The interfacial properties strongly affect chemi-cal, thermo-mechanical, and morphological prop-erties of the UV-exposed ZnO/PU films.
(3) By combining tapping mode AFM phase imaging and novel EFM phase imaging, the particle/ polymer interfacial regions can be successfully detected from the surface of the ZnO/PU films.
(4) The photocatalytic effect of ZnO nanoparticles is influenced by particle/polymer interfacial properties and the UV exposure environments. The ZnO nanoparticles can accelerate or reduce the photodegradation of polymeric coatings, depending on the chemistry of the coating system, the properties of nanoparticle surfaces, and expo-sure conditions (e.g., SPHERE vs UVO cleaner).
(5) The local interfacial regions are preferentially degraded when nanoparticles behave as a photo-catalyst, or protected when nanoparticles behave as a photostabilizer.
(6) The particle/polymer interface plays a critical role in photostability of nano-filled polymeric coatings.
(1.) Schadler, LS, Kumar, SK, Benicewicz, BB, Lewis, SL, Harton, SE, "Designed Interfaces in Polymer Nanocompositites: A Fundamental Viewpoint." MRS Bull., 32 335-340 (2007)
(2.) Gao, J, Zhao, B, Itkis, ME, Bekyarova, E, Hu, H, Kranak, V, Yu, A, Haddon, RC, "Chemical Engineering of the Single-Walled Carbon Nanotube-Nylon 6 Interface." J. Am. Chem. Soc, 128 7492-7496 (2006)
(3.) Gong, X, Liu, J, Baskaran, S, Voise, RD, Young, JS, "Surfactant-Assisted Processing of Carbon Nanotube/Poly-mer Composites." Chem. Mater., 12 1049-1052 (2000)
(4.) Calvert, P, "Nanotube Composites: A Recipe for Strength." Nature, 399 210-211 (1999)
(5.) Zhu, J, Kim, J. Peng, H, Margrave, JL, Khabashesku, VN, Barrera, EV, "Improving the Dispersion and Integration of Single-Walled Carbon Nanotubes in Epoxy Composites Through Functionalization." Nana Lett., 3 1107-1113 (2003)
(6.) Velasco-Santos, C, Martinez-Hernandez, AL, Fisher, FT, Ruoff, R, Castano, VM, "Improvement of Thermal and Mechanical Properties of Carbon Nanotube Composites Through Chemical Functionalization." Chem. Mater., 15 4470^475 (2003)
(7.) Akcora, P, et al., "Anisotropic Self-Assembly of Spherical Polymer-Grafted Nanoparticles." Nat. Mater., 8 354-359 (2009)
(8.) Nguyen. T, Granier, A, Steffens, C, Lee, H, Sharpiro, A, Martin, JW, "A Novel Method to Covalently Functionalize Carbon Nanotubes with Isocyanate for Polyurethane Nano-composite Coatings." Proc. ICE Coatings Tech Conference, Toronto, October 2007
(9.) Drzal, LT, "The Interface in Epoxy Composites." Adv. Polym. Sci, 75 1-32 (1986)
(10.) Drzal, LT, Rich, MJ, Koenig, MF, Lloyd, PF, "Adhesion of Graphite Fibers to Epoxy Matrices: 1. The Role of Fiber Surface Treatment." J. Adhes., 16 1-30 (1983)
(11.) Drzal, LT, Rich, MJ, Koenig, MF, Lloyd, PF, "Adhesion of Graphite Fibers to Epoxy Matrices. 2. The Effect of Fiber Finish." J. Adhes., 16 133-152 (1983)
(12.) Granick, S, et al., "Macromolecules at Surfaces: Research Challenges and Opportunities from Tribology to Biology." (13.) J. Polym. Sci. B: Polym. Phys., 41 2755-2793 (2003)
(13.) Jones, RL, Kumar, SK, Ho, DL, Briber, RM, Russell, TP, "Chain Conformation in Ultralhin Polymer Films." Nature, 400 146-149 (1999)
(14.) Frank, B, et al., "Polymer Mobility in Thin Films." Macro-molecules, 29 6531 (1996)
(15.) Ding, W, Eitan, A, Fisher, FT, Chen, X, Dikin, DA, Andrews, R, Brinson, LC, Schadler, LS, Ruoff, RS, "Direct Observation of Polymer Sheathing in Carbon Nanotube-Polycarbonate Composites." Nano Leu., 3 1593-1597 (2003)
(16.) Ciprari, D, Jacob. K, Tannenbaum, R. "Characterization of Polymer Nanocomposite Interphase and its Impact on Mechanical Properties." Macromolecules, 39 6565-6573 (2006)
(17.) Drzal, LT, "The Role of the Fiber-Matrix Interphase on Composite Properties." Vacuum, 41 1615-1618 (1990)
(18.) Li. X, el at, "Nanomechanical Characterization of Single-Walled Carbon Nanotube-Reinforced Epoxy Composites." Nanotechnology, 15 1416-1423 (2004)
(19.) Colling, JH. Dunderdale, J. "The Durability of Paint Films Containing Titanium Dioxide-Contraction, Erosion and Clear Layer Theories." Prog. Org. Coat., 9 47-84 (1981)
(20.) Clerici. C, Gu. X. Sung, LP. Forster, AM, Ho, DL, Stutzman. P, Nguyen. T, Martin, JW, "Effect of Pigment Dispersion on Durability of a Ti[O.sub.2] Pigmented Epoxy Coating during Outdoor Exposure," In: Martin, JW, Ryntz. RA, Chin, J, Dickie, RA (eds.) Service Life Prediction of Polymeric Materials: Global Perspectives, p. 475. Springer, New York. NY (2009)
(21.) Gu, X. et al., "Long-Term Performance of Nano-Filled Polymeric Materials: Effect of ZnO Nanoparticles on Pho-todegradation of A Waterborne Polyurethane Coating." Proc. CoatingsTech Conference, Indianapolis. IN. April 2009
(22.) Chin. J. Byrd, E. Embree, N, Garver, J, Dickens, B, Finn, T, Martin. J. "Accelerated UV Weathering Device Based on Intergrating Sphere Technology." Rev. Set Instrum., 75 4951-4959 (2004)
(23.) Chin. JW. Byrd, E. Embree, N, Martin, JW, Tate, JD, "Ultraviolet Chamber Based on Integrating Spheres for Use in Artificial Weathering." J. Coat. Technol., 74 39-44 (2002)
(24.) Zhao, M. Gu, X, Nguyen. T. "Surface and Subsurface Characterization of Nanostructures in Polymeric Coatings Using Quantitative Electric Force Microscopy." Proceedings, CoatingsTech Conference, Indianapolis, IN, April 2009
(25.) Bansal, A, et al., "Quantitative Equivalence Between Polymer Nanocomposites and Thin Polymer Films." Nat, Mater., 4 693 (2005)
(26.) Bansal, A. Yang. H, Li. C. Benicewicz. BC, Kumar. SK. Schadler, LS, "Controlling the Thermomechanical Properties of Polymer Nanocomposites by Tailoring the Polymer-Particle Interface." J. Polym. Sci. B: Polym., Phys., 44 2944-2950 (2006)
(27.) Ash. BJ, Schadler, LS, Siegel, RW, "Glass Transition Behavior of Alumina/Polymethylmethacrylate Nanocomposites," Mater. Lett., 55 (1-2) 83 (2002)
(28.) Starr. FW. Schroder. TB. Glotzer, SC "Molecular Dynamics Simulation of a Polymer Melt With a Nanoscopic Particle." Macromolecules, 35 4481-4492 (2002)
(29.) Oh. H, Green, PF, "Polymer Chain Dynamics and Glass Transition in a Thermal Polymer/Nanoparticle Mixtures." Nat. Mater.. 8 139-143 (2009)'
(30.) Gu. X, HNguyen. T. Oudina. M, Martin, D. Kidah. B. Jasmin. J. Rezig. A. Sung, LP. Byrd. E, Martin, JW, "Microstructure and Morphology of Amine-Cured Epoxy Coatings Before and After Outdoor Exposures-An AFM Study." J. Coat. Technol. Res.. 2 (7) 547-556 (2005)
(31.) Lin. F, Meier. DJ, "Latex Film Formation: Atomic Force Microscopy and Theoretical Results." Prog. Org. Coat.. 29 139-146 (1996)
(32.) Dobler, F, Pith. T, Lambla. M, Holl. Y. "Coalescence Mechanisms of Polymer Colloids. I. Coalescence Under the Influence of Particle-Water Interfacial Tension." J. Colloid Interface Sci, 152 111 (1992)
(33.) Dobler, F, Pith, T, Lambla, M, Holl, Y. "Coalescence Mechanisms of Polymer Colloids. II: Coalescence with Evaporation of Water." J. Colloid Interface Sci., 152 12-21 (1992)
(34.) Zheng, J, Ozisik, B, Siegel, RW. "Disruption of Self-assembly and Altered Mechanical Behavior in Polyurethane/Zinc Oxide Nanocomposites." Polymer, 46 10873-10882 (2005)
(35.) McLean, RS, Sauer, BB, "Tapping-Mode AFM Studies Using Phase Detection for Resolution of Nanophases in Segmented Polyurethanes and Other Block Copolymers." Macromolecules, 30 8314-8317 (1997)
(36.) Raghavan. D. VanLandingham, M, Gu, X, Nguyen. T. "Characterization of Heterogeneity in PMMA/PB Blends with Atomic Force Microscopy." Langmuir. 16 9448-9459 (2000)
(37.) Raghavan, D, Gu, X, Nguyen, T, VanLandingham. MR. Karim, A, "Mapping Polymer Heterogeneity Using Atomic Force Microscopy--Phase Imaging and Nanoscale Indentation." Macromotecules, 33 (7) 2573-2583 (2000)
(38.) Raghavan, D, Gu, X, Nguyen. T, VanLandingham, MR, "Characterization of Chemical Heterogeneity in Polymer Systems Using Hydrolysis and Tapping Mode Atomic Force Microscopy." J. Polym. Sci B: Polym. Phys.. 39 1460 1470 (2001)
(39.) Raghavan, D, Gu. X, Nguyen, T, VanLandingham. MR, "Mapping Chemically Heterogeneous Polymer System Using Selective Chemical Reaction and Tapping Mode Atomic Force Microscopy." Macromol. Symp., 167 297-305 (2001)
(40.) Nguyen, T, Jasmin. J, Sung, L. Gu, X. Rezig, A. Martin. D. Martin, JW, u Relation Between Chemical Degradation and Thickness Loss of Clear Crosslinked Polymeric Coatings Exposed to UV." In: Martin, JW. Ryntz, RA, Dickie, RA (eds.) Service Life Prediction: Challenge the Status Quo. p. 13. Federation of Societies for Coatings Technology, FSCT. Blue Bell, PA (2005)
(41.) VanLandingham, MR, Byrd, WE. Martin, JW, "On the Use of the Atomic Force Microscope to Monitor Physical Degradation of Polymeric Coating Surfaces." J. Coat. Technol., 73 43-50 (2001)
(42.) Lemaire. J. Siampiringue, N, "Prediction of Coating Lifetime Based on FTIR Microspectrophotometric Analysis of Chemical Evolutions." In: Bauer. DR, Martin, JW (eds.) Service Life Prediction of Organic Coatings: A Systems Approach, ACS Symposium Series 772. p. 198. American Chemical Society, Oxford Press, NY (1999)
(43.) Moustaghfir, A, et al., "Sputtered Zinc Oxide Coatings: Structural Study and Application to the Photoprotection of the Polycarbonate." Surf, Coal Technol., 180-181 642-645 (2004)
(44.) Hegedus, C. Pepe, F. Lindenmuth, D, Burgard, D, "Zinc Oxide Nanoparticle Dispersions as Unique Additives for Coatings." JCT CoatingsTech, April, 42-52 (2008)
(45.) Li, YQ, Fu, SY. Mai, YW, "Preparation and Characterization of Transparent ZnO/Epoxy Nanocomposites With High-UV Shielding Efficiency." Polymer, 47 2127-2132 (2006)
(46.) Yang. R, Li. Y, Yu, J, "Photo-Stabilization of Linear Low Density Polyethylene by Inorganic Nanoparticles." Polym. Degrad. Stab., 88 168-174 (2005)
(47.) Chandramouleeswaran, S, Mhaske, ST, Kalhe, AA, Varadarajan. PV, Prasad, V, Vigneshwaran. N. "Functional Behaviour of Polypropylene/ZnO-Soluble Starch Nanocom-posites." Nanotechnology, 18 385702 (2007)
(48.) Li, R, Zhao, H. "A Study of Photo-degradation of Zinc Oxide (ZnO) Filled Polypropylene Nanocomposites." Polymer. 47 3207-3217 (2006)
(49.) Baker, P, Branch, A, "The Interaction of Modern Sunscreen Formulation with Surface Coatings." Prog.Org. Coat., 62 313-320 (2008)
(50.) Yang, H, Zhu, S, Pan, N, "Studying the Mechanisms of Titanium Dioxide as Ultraviolet-Blocking Additive for Films and Fabrics by an Improved Scheme." J, Appl Polym. Sci, 92 3201-3210 (2004) 10.1007/s11998-011-9326-1
(51.) Xue, T, Nie, D, Zeng, S, Zhang, Y, Pan, L, "Preparation and Characterization of Coated ZnO Nanoparticles." Key Eng. Mater., 368-372 1636-1638 (2008)
J. Coat. Technol. Res., 9 (3) 251-267, 2012 DOI 10.10077/s11998-011-9326-1
* Instruments and materials are identified in this article to describe the experiments. In no case does such identification imply recommendation or endorsement by the National Institute of Standards and Technology (NIST).
X. Gu (*), G. Chen, M. Zhao, S. S. Watson, T. Nguyen, J. W. Chin, J. W. Martin
Materials and Construction Research Division, National Institute of Standards and Technology, Gaithersburg, MD 20899, USA